Skip to content
Menu

Stock Move Explainer

by Xbt LinSkillResearch something

Finds out why a stock jumped or dropped, with a dated event timeline and a likely cause

View on GitHub

The problem it solves

Your stock fell 9% this morning, the headlines contradict each other, and you need to know whether anything actually changed.

For Individual investorsPeople tracking stocks they own

Before
You flick between news sites, forums and filings and guess which story explains the drop.
After
You get a dated timeline across company, regulation, industry and sentiment, a cause table with evidence and confidence, and a note on whether to revisit your thesis.
Stars
★ 17k
License
MIT
Cost
Usage-based
Level
Intermediate
Contents
Includes scripts

Overview

This skill from the AI Berkshire collection answers one question fast: why did this stock just move? Give it a company and, if you like, the size of the move and a time window. Four parallel agents check company announcements, regulation, competitors and market sentiment, and the skill merges them into a dated timeline, a table of possible causes with evidence and confidence, and a verdict on whether the move looks like a real change or noise. The instructions and report are in Chinese. It suits investors who hold or watch a stock and want facts before reacting.

It is told to answer 'cause unclear' when the news does not match the size of the move, instead of forcing a story onto it.

Try these first

Start with the first one: it is a small task you can check easily.

  1. [Company] moved [+/- percent] on [date]. Check the last 7 days of news and link a source for every event.First try

  2. Is this move about [company] itself or the whole sector? Show how its peers moved.

  3. List what you could not find, and what would change your conclusion.

  4. Save the report to the reports folder.writes a file

Example prompts written by Agent Flows Hub.

What you get

From the creator's documentation.

OutputWhat it isReads
Four scoutsParallel agents for company events, regulation, peers and market sentimentWeb data
Event timelineOne dated, sourced timeline merged from all four checksYour agent
Cause tableCandidate explanations with evidence, counter-evidence, confidence and how long they may lastYour agent
VerdictReal change, sentiment, unclear or mixed, plus a follow-up list for the next 7 to 30 daysYour agent
Investor postsOptional pull of a well-known investor's Xueqiu posts, using your own loginYour Xueqiu account
Report fileThe report saved as a Markdown file in the repository's reports folderYour computer

Good to know

A stock you hold or watch moves sharply and you want to know why. The skill sends four agents through company news, regulation, peers and sentiment, then writes a dated timeline and a judgement on the main cause.

Limits

  • It starts four agents in parallel and reads many web pages. The optional step that pulls a well-known investor's Xueqiu posts needs Python, Playwright and your own Xueqiu login. The instructions and report are in Chinese.
  • The README states the project is for research and education, not investment advice; the skill lists follow-up actions but leaves every buy or sell decision to you.

Notes by (original project name: news-pulse (AI Berkshire)), written from the creator's documentation, not from running it ourselves. news-pulse SKILL.md · AI Berkshire README (English)

Stuck setting it up?

Tell us what happened. We take a look first; if it needs paid work, you get an estimate before anything starts.

Get help

More to research something

Other skills

See all 140 skills · A skill is installed but not loading

GitHub